Inferring causal networks from observations and interventions

By Mark Steyvers, Josh Tenenbaum, Eric-Jan Wagenmakers, Ben Blum. 2003

503 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Developmental and Educational Psychology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Developmental and Educational Psychology

Citations in the corpus

Discovering Causal Signals in Images (2016, background): “When such experiments are expensive, unethical, or impossible to conduct, we must discern cause from effect using observational data only, and without the ability to intervene [30].”
A theory of causal learning in children: causal maps and Bayes nets. (2004, methodology): “Tenenbaum and Griffiths (2003) have recently described a Bayesian Bayes net learning model for backward blocking with the data above.”
Preschool children learn about causal structure from conditional interventions. (2007, unknown): “…a variety of causal reasoning problems (Gopnik & Schulz, in press) both in adults (Glymour, 2001; Lagnado & Sloman, 2002; Rehder & Hastie, 2001; Steyvers, Tenenbaum, Wagenmakers & Blum, 2003; Tenenbaum & Griffiths, 2003; Waldmann & Hagmayer, 2001) and children (Gopnik, Sobel, Schulz & Glymour,…”
Theory-based causal induction. (2009, methodology): “This prediction has been evaluated by several different researchers (e.g., Gopnik et al., 2004; Hagmayer, Sloman, Lagnado, & Waldmann, 2007; Lagnado & Sloman, 2004; Steyvers et al., 2003).”
Structure learning in action (2010, background): “It is therefore not surprising that humans have been shown to use additional cues to identify causal dependencies like temporal order [52] and active interventions [47,50,51]—for example, by manipulating the barometer and testing whether this has an influence on the probability of a storm occurring.”
Formalizing Neurath’s Ship: Approximate Algorithms for Online Causal Learning (2016, background): “Likewise, adults and even children have been shown to select interventions that are robustly more informative than chance, but much less efficient than idealized active learning (Bramley, Lagnado, & Speekenbrink, 2015; Coenen, Rehder, & Gureckis, 2015; Gureckis & Markant, 2009; Lucas, Bridgers, Grif”
Learning a theory of causality. (2011, background): “…framework leads to quicker, more reliable learning than weaker assumptions about the nature of causation, and has been successful at predicting human learning data (Cheng, 1997; Waldmann & Martignon, 1998; Steyvers et al., 2003; Gopnik et al., 2004; Lu et al., 2008; Griffiths & Tenenbaum, 2005).”
Causal imprinting in causal structure learning. (2012, unknown): “For example, Steyvers et al. (2003) found that fewer than half of their participants could distinguish a common cause structure (A causes B and C) from a common effect structure (B and C cause A) based on observations alone.”
Bayesian models of cognition (2008, background): “…papers have investigated whether people are sensitive to the consequences of intervention, generally ﬁnding that people diﬀerentiate between observational and interventional evidence appropriately (Hagmayer, Sloman, Lagnado, & Waldmann, in press; Lagnado & Sloman, 2004; Steyvers et al., 2003).”
Motor Task Variation Induces Structural Learning (2009, unknown): “The principle of structural learning is not confined to the motor system and could also govern processes of perceptual and cognitive learning [31, 32].”
The propositional approach to associative learning as an alternative for association formation models (2009, background): “Few studies have looked at the latter issue (but see Steyvers et al., 2003).”
A glass-box interactive machine learning approach for solving NP-hard problems with the human-in-the-loop (2017, background): “More background on these problems can be found e.g. in (Chater et al., 2006), (Steyvers et al., 2003).”